#TTS model

KAN-TTS provides tools to develop custom text-to-speech models, supporting multiple languages ranging from Mandarin and English to regional dialects like Shanghainese and Cantonese. It employs models such as sam-bert and hifi-GAN for effective speech synthesis. Access training materials on the KAN-TTS Wiki and test examples via ModelScope. The platform is expanding with more languages and models. Community engagement is available through DingTalk.